Re: damned lies doesn't seem to be updating stats for accerciser

Yesterday at 11:35, Kenneth Nielsen wrote:

> 2007/9/9, Claude Paroz <claude 2xlibre net>:
>> Le dimanche 09 septembre 2007 à 13:00 +0200, Kenneth Nielsen a écrit :
>> > Hey I just had a look in Damned lies. Damned lies says that we have 1
>> > fuzzy in that translations. But when I check it out with svn (yes I
>> > checked this time that it is the same branch :$) and run
>> > intltool-update da and open it with an editor it shows it as being
>> > fully translated. It has been this way since september 1.
>> That seems to be related to the presence of an UTF-8 char in a msgid
>> ('(c)'). don't treat it correctly, hence the additional
>> string to translate.
>> I have two suppositions:
>> - gettext version on l.g.o is too old (0.14.5)
>> - there is a problem with UTF-8 encodings on the server
>> Maybe Danilo could give us some more hints about this problem.
>> BTW, I've seen a similar problem in comments in Orca po files.
>> Claude
> Ok thanks for clarifying that. We'll have to wait and see if anybody can
> help us with this then.

That's a bug in the application: Python assumes source code to be
Latin-1.  And xgettext recognizes it as such.  A .py file should use
'encoding: UTF-8' if it's directly embedding UTF-8 strings in a source

I reported a bug:


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]